Current melanoma treatments are either completely ineffective or they elicit short-lived responses, i.e., initial tumor shrinkage followed by rapid relapse. Our overall hypothesis is that therapy resistance in melanoma is due to a 'biologic signature', which is characterized through differential expression and activation of proteins in biologically defined sub-populations. Our working hypothesis is that chemo-resistance in melanoma is initially due to a defined subpopulation of malignant cells that evades antagonistic threats through slow or complete lack of proliferation. In addition, melanoma cells can develop acquired resistance mechanisms. In preliminary studies we have characterized a small population that cycles ten to hundredfold slower than the main population but that maintains high proliferation potential. This cell population, identified by the expression of the histone 3 K4 demethylase JARID1B, appears to serve as a tumor reservoir responsible for melanoma recurrence after therapeutic intervention as most chemotherapeutic agents and signaling inhibitors affect only rapidly dividing cells. We therefore hypothesize that melanomas can be effectively killed through a dual-tiered therapeutic approach: a genetic-based treatment aiming to eradicate the bulk of melanoma growths plus targeting of the remaining slow-cycling, self-renewing subpopulation. In Specific Aim 1, we will determine which signaling mechanisms slow-cycling self-renewing JARID1B+ cells apply to escape treatment with clinically used chemotherapeutic drugs and recently developed compounds targeting key signal transduction pathways. It is expected that slow-cycling, self-renewing JARID1B+ melanoma cells will remain largely unaffected whereas other populations are more sensitive. In Specific Aim 2, we plan to eliminate all cellular populations responsible for the malignant phenotype of advanced melanoma through a two-tiered therapeutic approach whereby the bulk of the population is killed through genetic-based targeted therapies and the slow-cycling subpopulation by a JARIDIB-targeting approach. These studies will enable the rational development of therapeutic modalities whereby oncologists can eliminate all melanoma cells and prevent drug resistance or recurrence.